AST SpaceMobile (ASTS)

Direct-to-handset broadband from LEO as MNO-partner infrastructure — the BlueBird-2 cadence over the next eight quarters decides $100bn TAM or zero.

Investment Research · Photoncap-style deep dive · Bucket B · 2026-05-22

What ASTS physically does

AST SpaceMobile is building a LEO constellation of unusually large satellites that act as cellular base-stations in the sky, communicating directly with unmodified standard mobile phones using terrestrial-MNO licensed spectrum. The technical mechanism is a very-large-aperture phased-array antenna — each BlueBird satellite presents an active aperture of roughly 64 m² in the first-generation BlueBird-1 design and approximately 223 m² in the second-generation Block-2 design, roughly an order of magnitude larger than any other commercial LEO satellite ever flown. The aperture size is not design vanity; it is dictated by link-budget physics. To close a link with a 0.5-watt mobile handset transmitter at 550-700km LEO altitude using cellular spectrum (700MHz to 2.6GHz primarily), the antenna gain has to compensate for ~90 dB of free-space path loss plus body loss, multipath, polarisation mismatch, atmospheric attenuation and interference margins. Smaller satellites cannot close that link with an unmodified phone.

The data-path is a 3GPP-compliant cellular waveform, not a proprietary protocol. To the handset, the satellite looks like a remote macro-cell — same air-interface, same authentication, same mobility-management procedures, with handover from terrestrial macro to satellite cell handled by the MNO core network and 3GPP Release-17 supplementary-coverage procedures. This is the structural difference from Starlink's direct-to-cell offering, which currently operates as a T-Mobile MVNO inside the Starlink mesh: Starlink controls the end-to-end stack and the MNO is effectively a wholesale buyer. ASTS by contrast is a wholesale-infrastructure supplier to the MNO, sharing revenue with them rather than disintermediating them — that is why AT&T, Verizon, Vodafone, Rakuten, Bell Canada, Telefonica, Orange, Smart Communications, MTN, Telkomsel, Millicom and Telstra have all signed commercial agreements.

The regulatory architecture matters: FCC Part 25 supplementary-coverage-from-space rules grant terrestrial-spectrum operators the right to provide satellite augmentation in their licensed band. The MNO-partner model is a regulatory entitlement, not just a commercial one — which is why MNOs strategically prefer ASTS to Starlink's vertical stack.

Product roadmap

BlueBird-1 (first-generation, ~64 m² aperture): five satellites launched in September 2024 on Falcon 9, already in commercial-grade service supporting voice, SMS and limited data under FCC special temporary authority. BlueBird-2 (Block-2, ~223 m² aperture): first batch of approximately five satellites targeted for Falcon 9 launch in Q2-Q3 2026, with an additional tranche later in 2026. Block-2 enables full 5G-NR support, target peak rates of 120 Mbps per beam with 30 Mbps average. Per January 2026 company update, the target is 17 satellites in service by end-2026, building toward a ~25-satellite operational coverage plane for equatorial markets in early 2027 and the full ~95-satellite global continuous-coverage constellation by 2028-2029.

Beyond Block-2: BlueBird-3 is in design with larger aperture, onboard L-band/S-band processing for direct DoD applications, and on-orbit-replenishment-friendly form factor. The Texas factory in Midland was expanded in 2024-2025 to support 6 satellites per quarter production cadence — the implied steady-state replacement rate for a 95-satellite constellation with 5-year design life. What ASTS deliberately does NOT contend in: backhaul-VSAT, GEO mobile satellite services, point-to-point microwave, terrestrial small-cells.

The financial print

Per the Q1 2026 print on May 13, ASTS reported revenue of approximately $12M (mostly gateway-test and government-development contracts), up from a near-zero base in Q1 2025. Quarterly operating loss was $63M with cash burn of approximately $190M including capex. FY2025 closed with revenue of approximately $5M and a full-year operating loss in the $250M range. Cash position at end-Q1 2026 was approximately $850M post the late-2025 convertible raise and Vodafone prepayment. Sell-side: Bernstein and Morgan Stanley carry "speculative buy" ratings with FY2026 consensus revenue of $40-90M driven by initial AT&T pilot service and government-development activations; FY2028 exit-rate consensus runs $700M-$1.4B depending on coverage assumptions. JPMorgan maintains a more cautious "neutral" with FY2027 revenue of $200-350M. Scotia initiated coverage in March 2026 at Sector Outperform with $135 PT.

The 1-year stock return through May 22 is approximately +285% (from ~$28 in May 2025 to $106.58 today). Q2 2026 earnings, expected mid-August, are the next binary — focal points are commercial-service tariff activation status with AT&T and BlueBird-2 launch-readiness milestones.

Customer mix today

The customer mix is structurally MNO-partner rather than direct-to-consumer. Per the company's 2025 disclosures, the named MNO partners cover a combined addressable subscriber base of approximately 2.8 billion handsets: AT&T (US, ~80M post-paid), Verizon (US, ~95M post-paid), Vodafone Group (multi-country, ~330M), Rakuten Japan (~6M), Bell Canada (~10M), Telefonica (multi-country, ~370M), Orange (multi-country, ~290M), Smart Communications Philippines (~70M), Saudi Telecom (~30M), MTN Africa (~290M), Telkomsel Indonesia (~160M), Millicom Tigo LatAm (~50M) and Telstra Australia (~20M). The economic model is a wholesale-rate-card for satellite-enabled cellular minutes and data, with ASTS receiving a share of MNO ARPU uplift — disclosed economics with Vodafone and Rakuten suggest a 50/50 split; AT&T and Verizon agreements are confidential but consensus assumes similar terms.

There is also a US government / DoD stream — a $43M Space Development Agency contract awarded in 2024 for direct-to-handset tactical applications, plus recurring Five-Eyes proof-of-concept work. The 2024-to-2026 shift is from "MNO partnerships announced but revenue zero" to "AT&T pilot revenue activated under FCC STA" by Q4 2025 to "AT&T commercial tariff filed and broader-MNO pilots activated" through 2026. That commercialisation step is the binary the whole equity story turns on.

What's actually happening at AT&T

AT&T is the most-advanced MNO partnership and the operational tell. The September 2024 announcement that AT&T would offer direct-to-handset emergency-text service via the first five BlueBirds was the first commercial-grade activation. Through 2025 the service expanded to two-way SMS, voice tests in select markets, and limited beta-data-service in late 2025 under FCC STA. The MNO commercial-launch milestone — the moment AT&T begins charging post-paid subscribers an incremental monthly fee for satellite-augmented coverage — is the catalyst that converts ASTS from "story stock" to "ratable-revenue space company".

AT&T's wireless segment runs ~$80bn annual revenue with 60-65% margins on a ~$200/year-per-sub basis. Even a $0.50/month upcharge on 80M post-paid subscribers is ~$480M annual AT&T-level revenue. ASTS economics are a 50/50 share of incremental ARPU, so $240M annual run-rate from AT&T alone is mathematically defensible if activation is meaningful. The planned commercial tariff is "Connectivity Anywhere," targeted to launch mid-to-late 2026 at $5-10/month per line. If 15% of post-paid subs adopt at $7.50/month, that's $1.08bn AT&T-segment annual revenue, $540M to ASTS at 50/50. If 5% adopt at $5/month, the math falls to $240M / $120M. The bull-case math depends entirely on adoption-rate validation in the first six months of commercial service — Q3 and Q4 2026 prints are the critical validation windows. The Apple-Globalstar emergency-SOS analogue suggests "always-on coverage" has meaningful consumer value.

The competitive threat / Starlink Direct-to-Cell, Lynk Global

The named competitor is Starlink Direct-to-Cell, now operationally live as a T-Mobile MVNO offering, with tens of millions of cells under it via the existing Starlink V2-mini constellation (over 7,000 satellites on-orbit, a meaningful subset V2-mini direct-to-cell capable). Starlink's advantage is the deployed-asset base and SpaceX's vertically-integrated launch economics. The disadvantage is the architecture: Starlink's direct-to-cell uses a software-defined cellular waveform on smaller satellites, with much lower per-cell capacity and a single MNO partner per market (T-Mobile in US, KDDI in Japan, Optus in Australia).

Lynk Global is private, operating a small constellation of similar-sized satellites with similar MNO-partner economics. Lynk has agreements in 50+ countries but has not raised the capital to deploy at ASTS's scale — last reported Series B was approximately $40M in 2024, materially below ASTS's $850M cash position. Iridium Certus is narrowband direct-to-handset but requires a special handset. Apple-Globalstar emergency-SOS is a feature not a service. The combined competitive picture: Starlink is the volume-leader at the lower-quality end, ASTS is the architecture-leader at the higher-quality end, Lynk is the capital-constrained third place.

The terminal risk

The structural risk is that direct-to-handset turns out to be a luxury feature with low willingness-to-pay rather than a coverage-gap-killer with high WTP. If MNOs find that fewer than 2-3% of post-paid subs pay a recurring fee for satellite augmentation, the revenue arithmetic collapses and ASTS becomes a zombie growth-stock that never reaches breakeven before exhausting capital. The secondary terminal risk is FCC spectrum-coordination friction — Verizon's CDMA/LTE legacy spectrum sits adjacent to ASTS operating bands; if the FCC requires deeper coordination or imposes power-flux limits, the link budget tightens and required satellite count rises. The tertiary risk is constellation-deployment-cost overrun: BlueBird-2 launches at roughly $30M per satellite all-in, and the company needs ~90+ for global operation — a ~$2.7bn capex bill against $850M cash. At least one major capital raise is mathematically required before constellation completion.

Bull / Gap / Optionality

Bull

1. AT&T commercial activation. The FCC special temporary authority granted in March 2025 permits commercial service rollout in non-rural geographies on a pilot basis; Q4 2025 commentary indicated a phased "thousands-then-millions" onboarding plan through 2026. If AT&T commercial service generates $240M run-rate by end-2026 at 50/50 split, ASTS revenue alone clears $120M against a current $12M Q1 run-rate.

2. European MNO activations. Vodafone has announced commercial-pilot intent for late 2026 in 4-5 markets, with Telefonica and Orange behind, providing a coordinated European commercial-launch cohort in 2027 that adds $200-400M incremental annual revenue at comparable adoption rates.

3. US DoD contract pipeline. SDA Tranche 3 procurement explicitly includes direct-to-handset capability and ASTS is the only commercial vendor with a flight-proven aperture at the relevant size — per Bernstein February 2026 note, the DoD-revenue ramp is plausibly $100-200M run-rate by 2027, with optionality for a $400-600M sole-source SBIRS-augmentation award in 2028-2029.

4. Launch-cadence step-up. Falcon 9 manifest commitments for 17 satellites across 2026 plus another tranche in 2027 means operational coverage materialises on a calendar-deterministic schedule, removing one of the historical execution risks (no hardware on-orbit for years).

5. Regulatory moat. FCC Part 25 supplementary-coverage-from-space rules grant terrestrial-spectrum operators the right to provide satellite augmentation in their licensed band, structurally favouring the MNO-partner model and disadvantaging a vertically-integrated competitor. That is the basis of AT&T's preference for ASTS over Starlink and is encoded in regulatory language that is durable rather than discretionary.

Gap

1. Cash-burn versus runway. $850M end-Q1 cash position against $190M quarterly burn means the company likely needs to raise capital by Q2 2027 unless commercial revenue inflects materially earlier. Equity raises at distressed valuations have historically been severely punitive — the 2022-2023 share-price action under capital-raise pressure was -75%.

2. Verizon-AT&T spectrum-coordination dispute. The September 2024 Verizon FCC filing flagged interference concerns that could constrain ASTS power-flux density limits, raising required satellite count for coverage and weakening unit economics. Verizon is now a partner BUT the legal residue remains.

3. BlueBird-2 deployment risk. Falcon 9 cadence is dependent on SpaceX manifest priority, where Starlink's own direct-to-cell sats are competing for the same fairing volume; a 6-9 month manifest slip materially postpones revenue and pushes any meaningful-cash-flow timeline into 2028.

4. Unit-economics opacity at end state. Management has guided to constellation steady-state operating margin of "30-50%" but has not provided per-customer ARPU economics; bears argue wholesale-MNO pricing power compresses to terrestrial-tower-rental-rate levels (15-20% sustainable margins, not 30-50%).

Optionality

The trade

ASTS at $106.58 sits at +27.7% vs 50MA with RSI 68.3 — hotter than ideal but not as stretched as RKLB on the same scan. The structural thesis is unique in public markets — there is no other listed direct-to-handset architecture play at this scale — and that scarcity supports a B-bucket rating despite the technical extension. Entry zone $101-$112 (current ±5%) for a 50bps starter, with the second 25bps tranche reserved for a pullback to the 50-day MA around $83.43. Sizing 75bps total, scaling to 150bps on confirmed AT&T commercial-service tariff activation in the Q2 or Q3 print. Stop: close below the 34/50 EMA cloud near $83 — that level coincides with the prior pivot consolidation; a break invalidates the BlueBird-2 deployment thesis. Catalyst: BlueBird-2 first-batch launch on Falcon 9, target Q3 2026 (July-September window). Pivot: if you want diversified launch + space-systems exposure rather than direct-to-handset specifically, RKLB above is the lower-beta compounder; ASTS is the high-beta direct-to-handset call option and should be sized accordingly.

Conviction: 7 / 10.

ticker: FLY name: Firefly Aerospace, Inc. theme: Space Aerospace bucket: B conviction: 5 entryzonelo: 46.07 entryzonehi: 50.91 currentprice: 48.49 pricedate: 2026-05-22 positionsizepct: 0.5 stoploss: 37.62 thesisoneline: Vertically-integrated launch + lunar-lander + responsive-space prime; the post-IPO catch-up to Rocket Lab if MLV hits 2026-2027. catalystnext: MLV first flight from Wallops catalystdate: 2026-12-15 rsi: 67.8 vs50ma: 28.9 forwardpe: 0.0 themecycleposition: early customermixsummary: Commercial launch ~40%, NASA CLPS ~25%, DoD/USSF ~25%, international/other ~10%. terminalriskoneline: MLV development slips past 2027 OR Alpha mission-failure recurrence — either obsoletes the medium-launch catch-up thesis. bulldriverscount: 5 gapriskscount: 4 optionalitycount: 6 lastearningsdate: 2026-05-15 nextearnings_date: 2026-08-14